What quantitative mechanical loading stimulates in vitro ...

Natenstedt et al. Journal of Experimental Orthopaedics (2015) 2:15 DOI 10.1186/s40634-015-0029-x

REVIEW Open Access

What quantitative mechanical loadingstimulates in vitro cultivation best?
Jerry Natenstedt1, Aimee C Kok2, Jenny Dankelman1 and Gabrielle JM Tuijthof1,2*
Abstract

Articular cartilage has limited regeneration capacities. One of the factors that appear to affect the in vitro cultivationof articular cartilage is mechanical stimulation. So far, no combination of parameters has been identified that offersthe best results. The goal is to review the literature in search of the best available set of quantitative mechanicalstimuli that lead to optimal in vitro cultivation.The databases Scopus and PubMed were used to survey the literature, and strict in- and exclusion criteria wereapplied regarding the presence of quantitative data. The review was performed by studying the type of loading(hydrostatic compression or direct compression), the loading magnitude, the frequency and the loading regime(duration of the loading) in comparison to quantitative evidence of cartilage quality response (cellular, signalingand mechanical).Thirty-three studies met all criteria of which 8 studied human, 20 bovine, 2 equine, 1 ovine, 1 porcine and 1 caninecells using four different types of cultivated constructs. Six studies investigated loading magnitude within the samesetup, three studies the frequency, and seven the loading regime. Nine studies presented mechanical tissueresponse. The studies suggest that a certain threshold exits for enhanced cartilage in vitro cultivation of explants(>20 % strain and 0.5 Hz), and that chondrocyte-seeded cultivated constructs show best results when loaded withphysiological mechanical stimuli. That is a loading pressure between 5–10 MPa and a loading frequency of 1 Hzexerted at intermittent intervals for a period of a week or longer. Critical aspects remain to be answered fortranslation into in vivo therapies.

Keywords: Chondrocytes; BMMSC; Compression; Mechanical loading; in vitro; Collagen type II; GAG; Cell therapy

IntroductionArticular cartilage is a nonlinearly permeable, viscoelasticmultiphasic material containing chondrocytes and proteo-glycan aggregates (3-10 % of volume) that are surroundedby an extracellular matrix (ECM), whose primary constitu-ents are water with mobile ions (60-85 % of volume) andcollagen type II (10-30 % of volume) (Fig. 1) (Mow et al.1999; Schulz and Bader 2007; Khan and Scott 2009;Madry et al. 2010). Damaged articular cartilage presentsitself as partial chondral, full thickness chondral or osteo-chondral defects (Fig. 1). Partial or full thickness lesionsshow limited ability to regenerate due to its avascular andhighly structured nature, which prevents progenitor cells

* Correspondence: [email protected] of Biomechanical Engineering, Faculty of Mechanical, Materialsand Maritime Engineering, Delft University of Technology, Mekelweg 2, Delft2628 CD, The Netherlands2Department of Orthopedic Surgery, Academic Medical Centre, Meibergdreef9, Amsterdam, AZ 1105, The Netherlands

© 2015 Natenstedt et al. This is an Open AccesLicense (http://creativecommons.org/licenses/bmedium, provided the original work is properly

and chondrocytes to migrate to the defect-site (Heath andMagari 1996; Zengerink et al. 2006; Khan et al. 2008;Magnussen et al. 2008; Khan and Scott 2009). For osteo-chondral defects, the subchondral bone plate is breachedleading to an inflow of blood containing bone marrow-derived mesenchymal stem cells (BMMSCs) that populatethe defect site (Khan and Scott 2009; Madry et al. 2011).These cells may differentiate into chondrocytic cells,which in turn can regenerate the ECM (Angele et al. 2003;Bahuleyan et al. 2009). This repair tissue mostly containscollagen type I and degrades over time (Khan et al. 2008;Madry et al. 2011; Hannon et al. 2014). However, newertissue engineered techniques are clinically applied inwhich tissue engineered constructs with or without(autologous) cells are used to enhance cartilage regener-ation with more hyaline like cartilage as result (Brittberg2010; Fortier et al. 2011; Hildner et al. 2011; Spiller et al.2011). Over the last decade, numerous studies have been

s article distributed under the terms of the Creative Commons Attributiony/4.0), which permits unrestricted use, distribution, and reproduction in anycredited.

http://crossmark.crossref.org/dialog/?doi=10.1186/s40634-015-0029-x&domain=pdf

mailto:[email protected]

http://creativecommons.org/licenses/by/4.0

Fig. 1 Cellular structure of cartilage. Defects are sustained in different layers: partial thickness chondral defect (up till the deep zone), full thicknesschondral defect (up till the calcified cartilage) and osteochondral defect (crossing the subchondral bone plate) base on (Madry et al. 2010)

Natenstedt et al. Journal of Experimental Orthopaedics (2015) 2:15 of 15

published that unravelled important factors influencingthe cartilage regeneration process (e.g. (Chung andBurdick 2008; Khan et al. 2008; Brittberg 2010; Fortieret al. 2011; Hildner et al. 2011; Spiller et al. 2011)).This review will focus on the mechanical loading to pro-

mote cartilage cultivation. Mechanical stimuli promoteECM production (Grodzinsky et al. 2000), increase chon-drocyte activity, and aid to protect the ECM temporarilyfrom excessive loading (Mow et al. 1999). Due to itsunique composition, cartilage can be loaded up to 18 MPain vivo, which is up to 15 times the body weight (Darlingand Athanasiou 2003; Elder and Athanasiou 2009; Spilleret al. 2011). If an underdeveloped ECM sustains such highloading, it can collapse; and can further deteriorate even-tually leading to a full stop in intrinsic repair (Darling andAthanasiou 2003). A similar mechanism is seen whendamaged cartilage (with a disrupted ECM) is loaded dur-ing gait with physiological values of around 5 times thebody weight (van Dijk et al. 2010a, 2010b).On the other hand, biomechanical intermittent cyclic

loading has shown to stimulate regeneration of cartilagetissue (Arokoski et al. 2000; Bonassar et al. 2001; Waldmanet al. 2004; Chung and Burdick 2008; Fan and Waldman2010; Hess et al. 2010; Potier et al. 2010). Tissue engineer-ing studies show that dynamic compression increases car-tilage in vitro cultivation rather than static compression

(Schulz and Bader 2007; Elder and Athanasiou 2008, 2009;Mizuno and Ogawa 2011). Important loading parametersappear to be the magnitude, frequency and duration(Ikenoue et al. 2003). So far, no combination of parametershas been identified that offers the best result in in vivoregeneration. Unfortunately, a first search in the literatureindicated that none of the retrieved in vivo studies pro-vided quantitative values to identify this combination.Therefore, the goal of this study is to review the literaturein search of the best available set of quantitative mec-hanical stimuli that increase cartilage in vitro cultivation;and possibly deduct suggestions for in vivo cartilageregeneration.

ReviewMethodsThe databases Scopus and PubMed where used to sur-vey the literature. The following keywords and synonymswere used to retrieve candidate studies: (Mechan* ORBiomech*) AND (load* OR loading OR stimulat* ORcompress* OR shear OR forces) AND articular cartilageAND (repair OR regeneration OR healing). Both originaland review papers were included from 1980 until April18th 2015 and the search was limited to English language.The in- and exclusion criteria were formulated based onstrict interpretation of the research question.


Inclusion criteriaStudies were included if they:

� Studied dynamic loading.� Harvested either chondrocyte or BMMSC cultures

from larger animal models (canine, bovine, equine,porcine, human). Larger animals provide a closerenvironment and metabolism to the human cartilagecase (Chu et al. 2010).

� Provided quantitative values of the appliedmechanical stimuli (loading magnitude, frequency,type and regime).

� Provided quantitative values of the effect on thecartilage quality response (e.g. cellular or mRNAresponse in terms of increased percentage).

Exclusion criteriaStudies were excluded if they only:

� Used computational methods to simulate themechanical properties or regenerating capabilitiesof cartilage.

� Determined the mechanical properties of cartilage(e.g. stiffness or elastic strain limit).

� Presented operative techniques to promote cartilagehealing (e.g. grafting and graft-ingrowth,arthroplasty, or microfracturing).

� Studied in vivo loading to stimulate cartilage, forexample by specific physical therapy protocols, anddid not provide quantitative data on the mechanicalstimuli and/or effect on the cartilage quality response.

� Studied non-articular cartilage (e.g. cartilage of the ear).� Examined the effect of non-mechanical factors

(e.g. hormonal, chemical, or electrical) that canstimulate cartilage cultivation or regeneration.

� Harvested chondrocytes or stem cells from smallanimals (e.g. rodent) (Chu et al. 2010).

� Performed continuous static compression asloading regime.

From the included studies, the cell donor and culture,the construct, the additives, the type of loading (hydro-static compression or direct compression), the loadingmagnitude, the frequency and the loading regime (dur-ation of the loading) were documented as input parame-ters, and the quantitative evidence of cartilage qualityresponse (cellular, signaling and mechanical responses)was documented as output parameter. Unfortunately,variety in testing conditions such as cell source, cultivatedconstructs, and the chosen cartilage quality response pa-rameters only allowed for a qualitative comparison. To en-hance comparison, the applied loading regime is presentedin one format: the frequency, the total time of loading perday, the total number of days and total loading hours.

Additionally, the loads were mostly expressed in pressure(P or σ equals force per area expressed in Pa) (Shepherdand Seedhom 1999; Darling and Athanasiou 2003). How-ever, some studies use the strain (ℇ), defined as the per-centage of cartilage thickness decrease. To enhanceuniformity, the strain values were converted into pressurevalues using the formula for linear elastic materials:

σ ¼ E⋅ε ð1Þ

Where E is the Young’s modulus, which is a materialproperty, and σ is the compression load expressed in Pa.Since cartilage is a viscoelastic multiphasic material(Mow et al. 1999), multiple parameters are needed to de-scribe its material behaviour. In this study, an approxima-tion of the Young’s modulus was used: the ‘instantaneous’compressive modulus of cartilage (Ec) (Shepherd andSeedhom 1999). Substituting Ec and ℇ in Equation (1)gives the compression load. The value of Ec depends onthe joint donor site (Shepherd and Seedhom 1999). Tothis end, the range of values for the human knee joint (be-tween 6 and 12 MPa) was filled out in Eq. (1) togetherwith the applied strain to calculate Ec for studies that usedthe human knee, the bovine and canine stifle joint asdonor sites (Shepherd and Seedhom 1999). In a similarfashion, the range of values for the human ankle (between11 and 19 MPa) was filled out for bovine or canine meta-tarsal donor joints (Shepherd and Seedhom 1999).Finally, this review presents the changes found in cellu-

lar, signaling and mechanical response due to the mechan-ical stimuli, which were indicated as an increased (+) ordecreased (−) response compared to controls or as nochange or similar (=) (Tables 1, 2, 3, 4, 5).

ResultsAfter a first title and abstract scan of the 836 hits com-bined from Scopus and PubMed, 106 papers were left tobe reviewed completely, which eventually resulted in 33studies that were eligible for inclusion. Generally, the tis-sue engineering studies investigated the effect of mechan-ical stimuli on cultivated constructs at least in one studyarm compared to unstimulated controls (Darling andAthanasiou 2003; Mauck et al. 2003). Tables 1, 2, 3, 4, 5summarize the results.

Cell sources, cultivated constructs and additivesTwo cell sources (chondrocytes and BMMSC) were stud-ied from 6 mammal types: 8 studies used human cells, 20bovine, 2 equine, 1 ovine, 1 porcine and 1 canine. First, 27studies used chondrocytes as cell source (Tables 1, 2, 3, 4).Chondrocytes are responsible for the production of theECM (Meachim and Stockwell 1973; Buckwalter andMankin 1998; Cohen et al. 1998), since they are likely tosynthesize collagen type II and proteoglycans (Schulz and

Table 1 Results on changes in cellular, signaling and/or mechanical response to explants for hydrostatic and direct compression (Fig. 2). PG is proteoglycan; MMP is matrixmetalloproteinases; # is number; h is hours; h/day is hours per day; − is decrease or inhibition; = is no change or status quo; + is increase; ++ is highest increase. *-symbolimplies pressure converted from strain, which is added in brackets

Study Cell source, cultivatedconstruct, additive(s)

Magnitude (MPa) Freq (Hz) Loading (h/day) Loading (# days) Loading (total h) Culture composition(change -, =, +)

mRNA response(change -, =, +)

Other findings(change -, =, +)

Hydrostatic compression

Parkinnen 1993(Parkkinen et al. 1993)

Bovine 5 0.5 1.5 1 1.5 + PG synthesis

Explant 0.25

Fetal calf serum 0.05 = PG synthesis

0.0167

Direct compression

Li 2013 (Li et al. 2013) Young Bovine BruisedExplant Serum free medium,20 g/ml ascorbic acid

0.6-1.2* (10 %) 0.5 4 4 16 + PG synthesis + aggrecan Control show betterresults compared tobruised explants+ collagen II

1.2-2.4* (20 %) ++ PG synthesis ++ aggrecan

++ collagen II

1.8-3.6* (30 %) + PG synthesis + aggrecan

= collagen II

Okuda 2013 (Okudaet al. 2013)

Young Bovine, Explant 20 %Fetal bovine serum; 50 mg/L L-ascorbic acid

0.6-1.2* (10 %) 1 3.5 5 17.5 + sGAG + compressivemodulus

+ # of cells

Torzilli 1996 (Torzilliet al. 1996)

Bovine, Explant 10 % Fetalbovine serum; 50 μg/mLascorbic acid

1 1 24 1 24 - PG synthesis

Torzilli 2011 (Torzilliet al. 2011)

Bovine, Explant 10 % Fetalbovine serum; 50 μg/mLascorbic acid

0.5 (10 %) 0.5 24 3 72 = PG content = MMP −3, −13

Natenstedt

etal.Journalof

ExperimentalO

rthopaedics (2015) 2:15

Page4of

15

Table 2 Results on changes in cellular, signaling and/or mechanical response to chondrocyte-seeded meshes for hydrostatic and direct compression (Fig. 2). PGA is polyglycolicacid; PEGT/PBT is polyethylene glycol terephthalate/polybutylene terephthalate; Sox9 is the gene that regulates chondrogenic differentiation; # is number; h is hours; h/day ishours per day; − is decrease or inhibition; = is no change or status quo; + is increase; ++ is highest increase. *-symbol implies pressure converted from strain, which is added inbrackets; **loading was performed every other day for 1 h twice a day with 8 h rest in between





Hydrostatic compression

Carver 1999a(Carver andHeath 1999a)

Young equine Mesh nonwovenPGA 10 % Fetal bovine serum;50 μg/mL ascorbic acid

3.4 0.25 2 35 70 + GAG GAG increasestrongest at 6.9 MPafor young= collagen II

= # of chondrocytes

6.9 + GAG

+ collagen II

= # of chondrocytes

Adult equine Mesh nonwovenPGA 10 % Fetal bovine serum;50 μg/mL ascorbic acid

3.4 + GAG Collagen II increasestrongest at 6.9 MPafor young and adult+ collagen II

= # of chondrocytes

6.9 = GAG

+ collagen II

= # of chondrocytes

Carver 1999b(Carver andHeath 1999b)

Young equine Mesh nonwovenPGA 10 % Fetal bovine serum;50 μg/mL ascorbic acid

3.44 0.25 2 35 70 + GAG + E- modulus

+ collagen

= # of chondrocytes

Direct compression

Démarteau2003(Démarteauet al. 2003)

Human Mesh PEGT/PBT Foam10 % Fetal bovine serum; growthfactor TGF-β1, FGF-2, PDGFbb

0.3-0.6* (5 %) 0.1 4 3 12 + GAG = Sox9 Measured peakloading 0.018 MPa

= aggrecan

= collagen II

Hilz 2014 (Hilzet al. 2014)

Bovine, Mesh Polyurethane 25 %Fetal calf serum;50 μg/mLL-ascorbic acid

1.2-2.4* (20 %) 1 2** 21 16 + GAG =Sox9

+ aggrecan +collagen II

+ collagen II

El-ayoubi 2011(El-Ayoubiet al. 2011)

Canine, Mesh poly-L-Lactide 10 %Fetal bovine serum

0.6-1.2* (10 %) 1 3 14 42 + # of cells

Natenstedt

etal.Journalof

ExperimentalO


Page5of

15

Table 3 Results on changes in cellular, signaling and/or mechanical response to chondrocyte cultivated constructs for hydrostatic compression (Fig. 2). PG is proteoglycan; MMPis matrix metalloproteinases; h is hours; h/day is hours per day;;- is decrease or inhibition; + is increase; ++ is highest increase; = is no change or status quo. *-symbol impliespressure converted from strain, which is added in brackets, **-symbol is increased aggrecan only with 4 h





Ikenoue 2003(Ikenoue et al.2003)

Human Monolayer 10 %Fetal bovine serum

1 1 4 1 4 = aggrecan Loading of 16h gavebetter resultscompared to 4h= collagen II

5 1 + aggrecan

10 = collagen II

1 1 4 4 16 + aggrecan

+ collagen II

5 1 + aggrecan

++ collagen II

10 1 ++ aggrecan

++ collagen II

Elder 2008 (Elderand Athanasiou2008)

Young bovine, Agarose gel,20 % Fetal bovine serum;50 μg/mL L-ascorbic acid

1 0.1 1 5 5 + GAG = Aggregatemodulus

= collagen II and #of cells

= E-modulus

5 0.1 = GAG + Aggregatemodulus


= E-modulus

10 0.1 + GAG = Aggregatemodulus


++ E-modulus

1 1 = GAG + Aggregatemodulus


+ E-modulus

5 1 = GAG = Aggregatemodulus


= E-modulus

10 1 ++ GAG ++ Aggregatemodulus


++ E-modulus

Natenstedt

etal.Journalof

ExperimentalO


Page6of

15

Table 3 Results on changes in cellular, signaling and/or mechanical response to chondrocyte cultivated constructs for hydrostatic compression (Fig. 2). PG is proteoglycan; MMPis matrix metalloproteinases; h is hours; h/day is hours per day;;- is decrease or inhibition; + is increase; ++ is highest increase; = is no change or status quo. *-symbol impliespressure converted from strain, which is added in brackets, **-symbol is increased aggrecan only with 4 h (Continued)

Hu 2006 (Huand Athanasiou2006)

Young bovine, Agarose gel, 10 %Fetal bovine serum; 50 μg/mL L-ascorbic acid

10 1 4 40 160 = GAG (no loss) +collagen II

= Aggregatemodulus

Mizuno 2011(Mizuno andOgawa 2011)

Young bovine, Aggregatepellet, Collagen solution, 10 %Fetal bovine serum

0.5 0.5 24 7 168 + sGAG + aggrecan, +collagen II, +MMP-3, −13

Kawanishi 2007(Kawanishiet al. 2007)

Young bovine, Aggregate pellet,10 % Fetal bovine serum; 50μg/mL ascorbic acid

5 0.5 4 4 16 + GAG + aggrecan

+ sGAG + collagen II

Suh 1999 (Suhet al. 1999)

Bovine, Monolayer, 10 % Fetalbovine serum

−0.08 vacuum 0.14 6 1 6 + PG synthesis + aggrecan

= collagen synthesis = collagen II

Parkinnen 1993(Parkkinenet al. 1993)

Bovine, Monolayer, 10 % Fetalcalf serum

5 0.5 1.5 1 1.5 - PG synthesis

0.25

0.05

0.0167 = PG synthesis

0.5 20 1 20 + PG synthesis

0.25

0.05 = PG synthesis

0.0167 - PG synthesis

0.0082 = PG synthesis

0.0034

Jortikka 2000(Jortikka et al.2000)

Bovine, Monolayer, 10 % Fetalbovine serum

5 0.5 20 1 20 + PG synthesis

Smith 1996(Smith et al.1996)

Bovine, Monolayer Ham’s F-12medium; 3 % Fetal bovine serum

10 1 4 1 4 + PG synthesis + aggrecan

+ collagen II

Smith 2000(Smith et al.2000)

Bovine, Monolayer, Ham’s F-12medium

10 1 2,4,8,12, 24 1 2,4,8,12, 24 + aggrecan**, +collagen II

Superior increasecompared to oneloading period

4 4 16 ++ aggrecan, ++ collagen II

Heyland 2006(Heyland et al.2006)

Porcine chondrocytes, Beadsalginate, 10 % Fetal bovineserum

0.3 0.0083 6 4 24 + GAG, = collagen II

6 7 42 = GAG, + collagen II = E-modulus

Natenstedt

etal.Journalof

ExperimentalO


Page7of

15

Table 4 Results on changes in cellular, signaling and/or mechanical response to chondrocyte cultivated constructs for direct compression (Fig. 2). OA is osteoarthritis; MMP ismatrix metalloproteinases; PG is proteoglycan; h is hours; h/day is hours per day; − is decrease or inhibition; + is increase; ++ is highest increase; = is no change or status quo.*-symbol implies pressure converted from strain, which is added in brackets,** -symbol is increase only present 12 h post stimulation, ^-symbol is increase only present after 6 hpost stimulation





Nebelung 2012(Nebelung et al.2012)

Human OA Hydrogel collagentype I 10 % Human serum

0.6-1.2* (10 %) 0.3 24 28 672 = proteoglycan = aggrecan = E-modulus

= collagen II + collagen II

+ MMP-13

Shelton 2003(Shelton et al.2003)

Bovine, Agarose gel Type VII20 % Fetal calf serum

1.7-2.9* (15 %) 0.3 24 2 48 - GAG

1 + GAG

3 = GAG

Omata 2012(Omata et al.2012)

Bovine, Agarose gel Type VII20 % Fetal bovine serum;0.85 mM L-ascorbic acid

1.7-2.9* (15 %) 1 6 22 132 + E-modulus

Hung 2004(Hung et al.2004)

Bovine, Agarose gel Type VII10 % Fetal bovine serum; growthfactor: TGF-β1, IGF-1

0.6-1.2* (10 %) 1 3 3 9 + aggrecan = aggregatemodulus

3 20 60 + E-modulus

+ aggregatemodulus

Nicodemus 2010(Nicodemus andBryant 2010)

Young bovine, Hydrogel polyethyleneglycol, 5 % Fetal bovine serum;50 mg/L L-ascorbic acid

1.2-2.4* (20 %) 0.3 24 7 168 + GAG + aggrecan

- collagen II

- MMP-3

= MMP-13

6 7 42 = GAG = aggrecan

+ collagen II

+ MMP-3, −13

Waldman 2004(Waldman et al.2004)

Bovine, Monolayer on top ofcalcium polyphosphate mesh,5 % Fetal bovine serum

0.3-0.6* (5 %) 1 .1 (400 cycles) 3.5 0.5 = PG synthesis

++ coll. synthesis

0.6-1.2* (10 %) ++ PG synthesis

1.2-2.4* (20 %) = coll. synthesis

0.3-0.6* (5 %) 0.6(2000 cycles)

3.5 2 = PG synthesis

+ coll. synthesis

0.6-1.2* (10 %) + PG synthesis

1.2-2.4* (20 %) = coll. synthesis

Natenstedt

etal.Journalof

ExperimentalO


Page8of

15

Table 4 Results on changes in cellular, signaling and/or mechanical response to chondrocyte cultivated constructs for direct compression (Fig. 2). OA is osteoarthritis; MMP ismatrix metalloproteinases; PG is proteoglycan; h is hours; h/day is hours per day; − is decrease or inhibition; + is increase; ++ is highest increase; = is no change or status quo.*-symbol implies pressure converted from strain, which is added in brackets,** -symbol is increase only present 12 h post stimulation, ^-symbol is increase only present after 6 hpost stimulation (Continued)

0.3-0.6* (5 %) 1 0.1 7 1 = PG synthesis = E-modulus

= coll. synthesis

0.1 14 2 + PG synthesis + E-modulus

+ coll. synthesis

De Croos 2006(De Croos et al.2006)

Bovine, Monolayer on top of calciumpolyphosphate mesh 5 % Fetal bovineserum

0.001 1 <1 h 1 <1 h + PG synthesis ^ + aggrecan **

+ coll. synthesis ^ + collagen II **

+ MMP-3, −13

Natenstedt

etal.Journalof

ExperimentalO


Page9of

15

Table 5 Results on changes in cellular and/or signaling response to BMMSC cultivated constructs under hydrostatic compression (Fig. 2). Sox9 is the gene that regulateschondrogenic differentiation; h is hours; h/day is hours per day; − is decrease or inhibition; and + is increase; ++ is highest increase; = is no change or status quo




Mesh

Wagner 2008 (Wagneret al. 2008)

Human BMMSC, Mesh CollagenType 1 50 mg/mL bovine serumalbumin; 50 μg/mL L-ascorbic acid;10−9 M dexamethasone

1 1 4 10 40 + proteoglycan + Sox9

+ aggrecan

+ collagen II

Luo 2007 (Luo andSeedhom 2007)

Ovine BMMSC, Mesh non-wovenfilamentous plasma-treated polyester10 % Fetal bovine serum; 50 μg/mLascorbic acid; 10−7 M dexamethasone

0.1 0.25 0.5 7 3.5 + GAG

= collagen

0.5 10 5 ++ GAG

+ collagen

Gel

Miyanishi 2006a (Miyanishiet al. 2006a)

Human BMMSC, Aggregate pellet1.25 mg/mL bovine serum albumin;50 μg/mL ascorbic acid; 10−7

M dexamethasone

0.1 1 4 14 56 = sGAG + Sox9

+ aggrecan

= collagen II

1 + sGAG ++ Sox9

+ aggrecan

= collagen II

10 ++ sGAG ++ Sox9

++ aggrecan

+ collagen II

Miyanishi 2006b (Miyanishiet al. 2006b)

Human BMMSC, Aggregate pellet1.25 mg/mL bovine serum albumin;50 cpg/mL ascorbic acid; 10−7

M dexamethasone

10 1 4 14 56 + Sox9

+ aggrecan

+ collagen II

Angele 2003 (Angeleet al. 2003)

Human BMMSC, Aggregate pellet10 % Fetal bovine serum

5.03 1 4 1 4 = proteoglycan

= collagen

4 7 28 + proteoglycan

+ collagen

Finger 2007 (Fingeret al. 2007)

Human BMMSC, Agarose gelType VII Growth medium Cambrex

7.5 1 4 14 56 = Sox9

Natenstedt

etal.Journalof

ExperimentalO


Page10

of15


Bader 2007; Spiller et al. 2011). Four different constructswere used to culture chondrocytes: a) explants, whichconsist of a complete section of cartilage that is excisedfrom a cadaver and embedded in a culture medium(Parkkinen et al. 1993) (Table 1), b) tissue engineeredmeshes that have a structural 3D shape (Table 2), c)monolayers that consist of isolated chondrocytes from fullthickness pieces of cartilage seeded onto a plate (Jortikkaet al. 2000; Smith et al. 2000) (Tables 3–4), d) hydrogelsthat have a softer structure compared to meshes (Carverand Heath 1999b; Hu and Athanasiou 2006) (Tables 3–4).One study used serum free medium (Li et al. 2013), onestudy added human serum (Nebelung et al. 2012), threeadded calf serum, and twenty one studies added bovineserum, with thirteen studies also adding L-ascorbic acidand two also adding growth factors in conjunction(Tables 1, 2, 3, 4).Second, BMMSCs were harvested from bone marrow,

and centrifuged to become a pellet culture (Miyanishiet al. 2006a; Kawanishi et al. 2007) (Table 5). Two culti-vated constructs were used onto which BMMSCs wereseeded : a) a gel or pellet composition and b) tissue engi-neered meshes (Luo and Seedhom 2007; Wagner et al.2008) (Table 5). Bovine serum was added in five out of 6studies, with four studies also adding 50 μg/mL L-ascorbicacid and dexamethasone.

Loading regimeTwo types of cyclic compression were applied: hydrostaticor direct compression (Fig. 2). Hydrostatic pressure isapplied by compressing the fluid surrounding the testedculture with a piston (Elder and Athanasiou 2009)(Tables 1, 2, 3 and 5). Direct compression implies that apiston directly presses on the tissue, which is commonlyexpressed in percentage of strain (Demarteau et al. 2003)(Tables 1,2 and 4). Except the studies by Torzelli et al.

Fig. 2 Difference between hydrostatic compression (left) and direct compr

(Torzilli et al. 1996; Torzilli et al. 2011) and De Croos et al.(De Croos et al. 2006), all other eleven studies requiredconversion from strain to pressure (Tables 1, 2 and 4).The loading magnitude, frequency and regime varied

highly. For example Carver and Heath stimulated theirsamples with 6.9 MPa at 0.25Hz for 2 h per day over aperiod of 35 days (Carver and Heath 1999b), whileDémarteau et al. applied a loading of 5 % of strain with0.1Hz for 4 h per day over a period of 3 days (Démarteauet al. 2003). In contrast to this, five out of six studies withhuman derived BMMSC used 1 Hz as loading frequencyfor 4 h per day (Table 5) (Angele et al. 2003; Miyanishiet al. 2006a, 2006b; Finger et al. 2007; Wagner et al. 2008).

Cartilage cellular, signaling and mechanical responseparametersThree types of methods were found to document cartilagequality response: cellular, signaling, and mechanical re-sponses (Tables 1, 2, 3, 4, 5). Cellular response is routinelydetermined with histology, which allows identification ofspecific (macro) molecules that typically represent healthycartilage (proteoglycans, glucosaminoglycans (GAGs) andsulfated glucosaminoglycans (sGAGs), collagen type II).An increase of proteoglycans is typically determined usingstaining with Safranin O (Darling and Athanasiou 2003;Schulz and Bader 2007). Similarly, the increase of GAGsand sGAGs is determined with dimethylmethylene blueassay staining (Farndale et al. 1986; Carver and Heath1999b; Shelton et al. 2003; Heyland et al. 2006; Hilz et al.2014). The increase in collagens is determined by stainingwith Picrosirius red, Masson’s trichrome stain, orantibody-staining such as anti-collagen antibodies ormonoclonal antibodies, and/or the use of polarized light(Angele et al. 2003; Darling and Athanasiou 2003; Heylandet al. 2006; Elder and Athanasiou 2008; Nicodemusand Bryant 2010). After staining, the histologic samples

ession (right). The arrows indicate the loading direction


are further interpreted with histologic scores andcompared to control samples to indicate relative cellularresponses. The type of collagen is assessed withimmunohistochemistry (Elder and Athanasiou 2008). Fi-nally, cellular response in the form of proteoglycan syn-thesis is routinely determined by assessing the radioactivelabeled 35Sulfate-uptake by the proteoglycans (Parkkinenet al. 1993; Torzilli et al. 2011; Li et al. 2013).Signaling response indirectly indicates the potential of

the cells to (de)differentiate into cartilage, because itassesses changes in the level of mRNA expression asproduced by the chondrocyte cells with a reverse tran-scription polymerase chain reaction (RT-PCR) (Darlingand Athanasiou 2003; Schulz and Bader 2007). A changein proteoglycan production is commonly documented asan increase in aggrecan mRNA that is the core proteinbackbone to which GAGs chains are attached (Tables 1,2, 3, 4) (Démarteau et al. 2003; Ikenoue et al. 2003;Schulz and Bader 2007). With the same method otherrelevant expressions are assessed: collagen type II, thegene sex determining region box 9 (Sox9) and Matrixmetalloproteinases (MMP). Sox9 is indicative for theregulation of chondrogenic differentiation and plays arole in expression of collagen type II and aggrecan(Démarteau et al. 2003; Miyanishi et al. 2006a; Fingeret al. 2007; Wagner et al. 2008; Hilz et al. 2014). MMPplays a major role in ECM turnover and degradation(Hilz et al. 2014). MMP-3 has shown to be a key playerin degrading matrix and inactivating other degradingenzymes (Cawston and Wilson 2006; Echtermeyer et al.2009), and MMP-13 seems to influence the progress ofosteoarthritis (Hilz et al. 2014).Mechanical responses were documented by compres-

sion tests of the samples and determining the Young’s(or E-) modulus from the linear range of the stress–strain curve (Carver and Heath 1999b; Hung et al. 2004;Heyland et al. 2006; Hu and Athanasiou 2006; Elder andAthanasiou 2008; Omata et al. 2012). Cartilage qualityresponse was given as change of cellular response(including synthesis) by 27 of 33 studies, as change ofsignaling response by 21 of 33 studies, and as change ofmechanical response by 9 of 33 studies.

Responses to mechanical stimuliDue to the difference in cell sources, cultivated constructs,compression and outcome measures only qualitative com-parison could be performed. Bovine explant testingmimics the in vivo case most closely especially the studyby Li et al. (Li et al. 2013), since they used bruised explants(Table 1). The studied variations in the loading regime ofthe explants suggest a certain threshold for the magnitudeof loading (>10 % or > 1.2 MPa) and the frequency(>0.5 Hz) to stimulate increased proteoglycan synthesis(Table 1). Only Okuda et al. (Okuda et al. 2013) confirmed

that this was correlated to increased mechanical response.Studies that tested equine chondrocyte-seeded meshes byhydrostatic compression support the need for a loadingthreshold (Table 2), since increased cell density wasobserved by higher loading magnitudes and correlated toincreased mechanical response (Carver and Heath 1999b).The three studies of chondrocyte-seeded meshes by directcompression presented too much variation to point in aspecific loading regime (Table 2).The studies that use monolayer and gel cultivated con-

structs and test variations in loading magnitude, frequencyor loading regime (Tables 3–4) also suggest the need forthresholds in magnitude and frequency with a trend to-wards higher values (up to 5–10 MPa and up to 1 Hz)compared to the explant studies to achieve increased re-sponses (Parkkinen et al. 1993; Smith et al. 2000; Ikenoueet al. 2003; Shelton et al. 2003; Waldman et al. 2004; Elderand Athanasiou 2008). Tables 3–4 also highlight the effectof different loading regimes, which seems to suggest thatprolonged duration and loading at intervals (no continu-ous intermittent loading) increase cell density and synthe-sis, signaling response as well as mechanical response (e.g.(Shelton et al. 2003; Hung et al. 2004)). Exceptions are thestudies by De Croos (De Croos et al. 2006) who find in-creased response at a low magnitude of 0.001 MPa appliedfor less than 1 h, and by Hu and Athanasiou (Hu andAthanasiou 2006) who did not find a change in theE-modulus after prolonged loading for 160 h.The studies using BMMSCs as cell source support the

suggestion that the largest increase in cellular and signal-ing response is achieved for larger loading magnitudes(>5 MPa) at a frequency of 1 Hz for a prolonged period(>7 days) at intermittent intervals (Table 5). However, nomechanical responses were measured for these constructs.

DiscussionFive studies mimicked the in vivo case most closely bytesting bovine explants. Only two of these varied the load-ing magnitude or frequency, which suggest the need of acertain threshold (>20 % strain and > 0.5 Hz) for increasedproteoglycan synthesis (Table 1). A careful qualitative in-terpretation of the results suggests that for chondrocyte-seeded cultivated constructs a loading pressure between5–10 MPa and a loading frequency of 1 Hz exerted atintermittent intervals for a period of a week or longer arerecommended as appropriate mechanical stimulus. Thesevalues are in the physiologic range of normal gait (Waterset al. 1988; Giddings et al. 2000; Brand 2005; Doke et al.2005; van Dijk et al. 2010a). Due to the variety of testingconditions and methods to express cartilage quality re-sponse, only qualitative comparison was possible, whichposes limitations to the study. First, differences in sampletissue, sample preparation, donor type and donor age allaccounted for differences in the outcome of these studies


(Parkkinen et al. 1993; Carver and Heath 1999a; Darlingand Athanasiou 2003; Chung and Burdick 2008). Still,fourteen of the 33 studies did measure the cartilage qualityresponse to varying loading parameters within the sameset up. Even though, signaling response does not alwaysreflect actual cellular and mechanical changes, the studiesthat report them also report cellular and/or mechanicalresponse in conjunction (Tables 2, 3, 4, 5), with in the ma-jority of the cases showing corresponding in- or decreases.With this, the decision was made to include studies thatonly present signaling response (four in total), since twohave varied the loading regime (intermittent vs continuousloading (Smith et al. 2000); and loading magnitude andduration (Ikenoue et al. 2003)) as needed to answer ourmain research question (Table 3). Second, two differenttypes of compression were applied: hydrostatic (Tables 1,2, 3 and 5) and direct compression (Tables 1,2 and 4).There is an on-going debate which type of loading is morephysiological. Some are in favor of direct compression(Buschmann et al. 1995; Mauck et al. 2000; Waldmanet al. 2004). Also, direct compression allows continuousmeasurement of mechanical responses, but needs sometricks to be applied to soft constructs by placing the sam-ples in bags (De Croos et al. 2006). Bachrach et al.(Bachrach et al. 1998) suggest that hydrostatic pressureseems to be more representative for the in vivo loadingcase, because it mimics the viscoelastic multiphasic cartil-age behavior closest. A drawback is that it also stimulatesthe sides of the samples. An advantage of applying hydro-static compression is that it allows for harmonization ofthe applied load, and it allows mechanical stimulation ofdifferent types of cultivated constructs including the softerones. The proposed transformation procedure from strainto pressure seems to make sense, because the calculatedpressure values are in line with the values found in otherstudies: 3.6 MPa leads to a 29 % strain (Herberhold et al.1999) vs 20 % strain (Hilz et al. 2014) (Tables 2 and 4).However, it still remains an approximation, which needsto be interpreted with care. Third, biologic demonstrationof the increase in cartilage quality response is highly im-portant, since it indicates parameters (signals, cells types,cell synthesis) that should be triggered to stimulate thecell activity and behave like cartilage. However, documen-tation of actual mechanical response would be expected aswell, since this determines performance. In one quarter ofthe studies (9 out of 33) the mechanical response wasmeasured, which is a rather low percentage. Some of thecultivated constructs (monolayer, pellet) do not resemblethe actual ECM structure, which makes mechanical test-ing difficult or impossible (Tables 1, 2, 3, 4, 5). Fullcharacterization is difficult, because of its highly complexviscoelastic behavior (Mow et al. 1999; Schulz and Bader2007; Khan and Scott 2009; Madry et al. 2010). Further-more, constructs can also change due to the loading or do

not necessarily mimic mechanical cartilage behavior(Nebelung et al. 2011). This latter is supported by conflict-ing results that were found for two studies in which agar-ose gel was used: Hu and Athanasiou (Hu and Athanasiou2006) show that a 20 % increase in collagen type II doesnot seem to influence the mechanical properties (Table 3),and others (Hung et al. 2004; Elder and Athanasiou 2008;Omata et al. 2012) did not find a relation between histo-logic and mechanical parameters.In vivo tissue engineering cartilage repair techniques

(e.g. Matrix-Induced Autologous Chondrocyte Implant-ation or cell-seeded hydrogel plugs (Brittberg 2010;Fortier et al. 2011; Hildner et al. 2011; Spiller et al.2011)) make use of similar scaffolds. This review gives asummary of current evidence, which can be used for fu-ture development of on in vivo application rehabilitationprotocol. Several factors are fundamentally different forthe in vivo case, including the fact that the ECM is notintact as result of the cartilage lesion, and that the accessto essential biologicals (e.g. cytokines, growth factors) isdifferent in the physiologic situation compared to thein vitro situation. Especially, the boundary between thehealthy cartilage and tissue engineered scaffold is a vul-nerable spot (Khan et al. 2008), which most likely cannotwithstand the suggested loading magnitude (Guettleret al. 2004; Khan et al. 2008; van Dijk et al. 2010a,2010b; Spiller et al. 2011; Hunt et al. 2012). However,the results could be used to optimize preconditioning oftissue engineered scaffolds before implantation into pa-tients (Shelton et al. 2003; Nebelung et al. 2012; Omataet al. 2012). Therefore, the timing of loading could be acritical factor that needs to be further explored. For ex-ample the testing period might be even further extended(Waldman et al. 2004), since in vivo studies with animalmodels evaluated the cartilage quality response afterlong testing periods (56 days or longer), much longerthan those found in this review (Saris et al. 2003; Koket al. 2013; Miller et al. 2014; Ortved et al. 2015). Finally,studying the dynamic compression of damaged explants(e.g. (Li et al. 2013)), should be elaborated to identify thebest magnitude, frequency and loading regime, sincethese constructs mimic the in vivo cartilage lesion closest.This will facilitate the translation of the found combin-ation of mechanical parameters to patients.

ConclusionsOverall, the results seem to suggest that a certain thresh-old exits for enhanced cartilage in vitro cultivation of ex-plants, and that chondrocyte-seeded cultivated constructsshow best results when loaded with physiological mechan-ical stimuli. This seems a reasonable conclusion, becausenature is highly optimized for daily activities such as nor-mal walking. Critical aspects remain to be answered fortranslation of the results into in vivo therapies.


Competing interestsThe authors declare that they have no competing interests.

Authors’ contributionsJN, GT: have made substantial contributions to conception and design, oracquisition of data, or analysis and interpretation of data; JN, AK, JD, GT: havebeen involved in drafting the manuscript or revising it critically for importantintellectual content; AK, JD, GT: have given final approval of the version tobe published.

Received: 7 April 2015 Accepted: 26 May 2015

ReferencesAngele P, Yoo JU, Smith C, Mansour J, Jepsen KJ, Nerlich M, Johnstone B (2003)

Cyclic hydrostatic pressure enhances the chondrogenic phenotype of humanmesenchymal progenitor cells differentiated in vitro. J Orthop Res 21(3):451–457

Arokoski JP, Jurvelin JS, Vaatainen U, Helminen HJ (2000) Normal andpathological adaptations of articular cartilage to joint loading. Scand J MedSci Sports 10(4):186–198

Bachrach NM, Mow VC, Guilak F (1998) Incompressibility of the solid matrix ofarticular cartilage under high hydrostatic pressures. J Biomech 31(5):445–451

Bahuleyan B, Cheung HS, Huang CYC (2009) Role of biomechanical force in stemcell-based therapy for cartilage repair. Curr Rheumatol Rev 5(1):34–39

Bonassar LJ, Grodzinsky AJ, Frank EH, Davila SG, Bhaktav NR, Trippel SB (2001)The effect of dynamic compression on the response of articular cartilage toinsulin-like growth factor-I. J Orthop Res 19(1):11–17

Brand RA (2005) Joint contact stress: a reasonable surrogate for biologicalprocesses? Iowa Orthop J 25:82–94

Brittberg M (2010) Cell carriers as the next generation of cell therapy for cartilagerepair: a review of the matrix-induced autologous chondrocyte implantationprocedure. Am J Sports Med 38(6):1259–1271

Buckwalter JA, Mankin HJ (1998) Articular cartilage: tissue design andchondrocyte-matrix interactions. Instr Course Lect 47:477–486

Buschmann MD, Gluzband YA, Grodzinsky AJ, Hunziker EB (1995) Mechanicalcompression modulates matrix biosynthesis in chondrocyte/agarose culture.J Cell Sci 108(Pt 4):1497–1508

Carver SE, Heath CA (1999a) Increasing extracellular matrix production inregenerating cartilage with intermittent physiological pressure. BiotechnolBioeng 62(2):166–174

Carver SE, Heath CA (1999b) Influence of intermittent pressure, fluid flow, andmixing on the regenerative properties of articular chondrocytes. BiotechnolBioeng 65(3):274–281

Cawston TE, Wilson AJ (2006) Understanding the role of tissue degradingenzymes and their inhibitors in development and disease. Best Pract Res ClinRheumatol 20(5):983–1002

Chu CR, Szczodry M, Bruno S (2010) Animal models for cartilage regenerationand repair. Tissue Eng Part B Rev 16(1):105–115

Chung C, Burdick JA (2008) Engineering cartilage tissue. Adv Drug Deliv Rev60(2):243–262

Cohen NP, Foster RJ, Mow VC (1998) Composition and dynamics of articularcartilage: structure, function, and maintaining healthy state. J Orthop SportsPhys Ther 28(4):203–215

Darling EM, Athanasiou KA (2003) Biomechanical strategies for articular cartilageregeneration. Ann Biomed Eng 31(9):1114–1124

De Croos JN, Dhaliwal SS, Grynpas MD, Pilliar RM, Kandel RA (2006) Cycliccompressive mechanical stimulation induces sequential catabolic andanabolic gene changes in chondrocytes resulting in increased extracellularmatrix accumulation. Matrix biology : journal of the International Society forMatrix Biology 25(6):323–331

Demarteau O, Jakob M, Schafer D, Heberer M, Martin I (2003) Development andvalidation of a bioreactor for physical stimulation of engineered cartilage.Biorheology 40(1–3):331–336

Démarteau O, Wendt D, Braccini A, Jakob M, Schäfer D, Heberer M, Martin I(2003) Dynamic compression of cartilage constructs engineered fromexpanded human articular chondrocytes. Biochem Biophys Res Commun310(2):580–588

Doke J, Donelan JM, Kuo AD (2005) Mechanics and energetics of swinging thehuman leg. J Exp Biol 208(Pt 3):439–445

Echtermeyer F, Bertrand J, Dreier R, Meinecke I, Neugebauer K, Fuerst M, Lee YJ,Song YW, Herzog C, Theilmeier G, Pap T (2009) Syndecan-4 regulates

ADAMTS-5 activation and cartilage breakdown in osteoarthritis. Nat Med15(9):1072–1076

El-Ayoubi R, DeGrandpre C, DiRaddo R, Yousefi AM, Lavigne P (2011) Design anddynamic culture of 3D-scaffolds for cartilage tissue engineering. J BiomaterAppl 25(5):429–444

Elder BD, Athanasiou KA (2008) Synergistic and additive effects of hydrostaticpressure and growth factors on tissue formation. PLoS One 3(6), e2341

Elder BD, Athanasiou KA (2009) Hydrostatic pressure in articular cartilage tissueengineering: from chondrocytes to tissue regeneration. Tissue EngPart B Rev 15(1):43–53

Fan JC, Waldman SD (2010) The effect of intermittent static biaxial tensile strainson tissue engineered cartilage. Ann Biomed Eng 38(4):1672–1682

Farndale RW, Buttle DJ, Barrett AJ (1986) Improved quantitation anddiscrimination of sulphated glycosaminoglycans by use ofdimethylmethylene blue. Biochim Biophys Acta 883(2):173–177

Finger AR, Sargent CY, Dulaney KO, Bernacki SH, Loboa EG (2007) Differentialeffects on messenger ribonucleic acid expression by bone marrow-derivedhuman mesenchymal stem cells seeded in agarose constructs due toramped and steady applications of cyclic hydrostatic pressure. TissueEng 13(6):1151–1158

Fortier LA, Barker JU, Strauss EJ, McCarrel TM, Cole BJ (2011) The role of growthfactors in cartilage repair. Clin Orthop Relat Res 469(10):2706–2715

Giddings VL, Beaupre GS, Whalen RT, Carter DR (2000) Calcaneal loading duringwalking and running. Med Sci Sports Exerc 32(3):627–634

Grodzinsky AJ, Levenston ME, Jin M, Frank EH (2000) Cartilage tissue remodelingin response to mechanical forces. Annu Rev Biomed Eng 2:691–713

Guettler JH, Demetropoulos CK, Yang KH, Jurist KA (2004) Osteochondral defectsin the human knee: influence of defect size on cartilage rim stress and loadredistribution to surrounding cartilage. Am J Sports Med 32(6):1451–1458

Hannon CP, Smyth NA, Murawski CD, Savage-Elliott I, Deyer TW, Calder JD,Kennedy JG (2014) Osteochondral lesions of the talus: aspects of currentmanagement. The bone & joint journal 96-B(2):164–171

Heath CA, Magari SR (1996) Mini-review: Mechanical factors affecting cartilageregeneration in vitro. Biotechnol Bioeng 50(4):430–437

Herberhold C, Faber S, Stammberger T, Steinlechner M, Putz R, Englmeier KH,Reiser M, Eckstein F (1999) In situ measurement of articular cartilagedeformation in intact femoropatellar joints under static loading. J Biomech32(12):1287–1295

Hess R, Douglas T, Myers KA, Rentsch B, Rentsch C, Worch H, Shrive NG, Hart DA,Scharnweber D (2010) Hydrostatic pressure stimulation of humanmesenchymal stem cells seeded on collagen-based artificial extracellularmatrices. J Biomech Eng 132(2):021001

Heyland J, Wiegandt K, Goepfert C, Nagel-Heyer S, Ilinich E, Schumacher U,Portner R (2006) Redifferentiation of chondrocytes and cartilage formationunder intermittent hydrostatic pressure. Biotechnol Lett 28(20):1641–1648

Hildner F, Albrecht C, Gabriel C, Redl H, van Griensven M (2011) State of the artand future perspectives of articular cartilage regeneration: a focus onadipose-derived stem cells and platelet-derived products. J Tissue Eng RegenMed 5(4):e36–51

Hilz FM, Ahrens P, Grad S, Stoddart MJ, Dahmani C, Wilken FL, Sauerschnig M,Niemeyer P, Zwingmann J, Burgkart R, von Eisenhart-Rothe R, Südkamp NP,Weyh T, Imhoff AB, Alini M, Salzmann GM (2014) Influence of extremely lowfrequency, low energy electromagnetic fields and combined mechanicalstimulation on chondrocytes in 3-D constructs for cartilage tissue engineering.Bioelectromagnetics 35(2):116–128

Hu JC, Athanasiou KA (2006) The effects of intermittent hydrostatic pressure onself-assembled articular cartilage constructs. Tissue Eng 12(5):1337–1344

Hung CT, Mauck RL, Wang CC, Lima EG, Ateshian GA (2004) A paradigm forfunctional tissue engineering of articular cartilage via applied physiologicdeformational loading. Ann Biomed Eng 32(1):35–49

Hunt KJ, Lee AT, Lindsey DP, Slikker W 3rd, Chou LB (2012) Osteochondral lesionsof the talus: effect of defect size and plantarflexion angle on ankle jointstresses. Am J Sports Med 40(4):895–901

Ikenoue T, Trindade MCD, Lee MS, Lin EY, Schurman DJ, Goodman SB, Smith RL(2003) Mechanoregulation of human articular chondrocyte aggrecan andtype II collagen expression by intermittent hydrostatic pressure in vitro.J Orthop Res 21(1):110–116

Jortikka MO, Parkkinen JJ, Inkinen RI, Karner J, Jarvelainen HT, Nelimarkka LO,Tammi MI, Lammi MJ (2000) The role of microtubules in the regulation ofproteoglycan synthesis in chondrocytes under hydrostatic pressure. ArchBiochem Biophys 374(2):172–180


Kawanishi M, Oura A, Furukawa K, Fukubayashi T, Nakamura K, Tateishi T, UshidaT (2007) Redifferentiation of dedifferentiated bovine articular chondrocytesenhanced by cyclic hydrostatic pressure under a gas-controlled system.Tissue Eng 13(5):957–964

Khan IM, Gilbert SJ, Singhrao SK, Duance VC, Archer CW (2008) Cartilageintegration: evaluation of the reasons for failure of integration duringcartilage repair. A review. Eur Cell Mater 16:26–39

Khan KM, Scott A (2009) Mechanotherapy: how physical therapists’ prescriptionof exercise promotes tissue repair. Br J Sports Med 43(4):247–252

Kok AC, Tuijthof GJ, den Dunnen S, van Tiel J, Siebelt M, Everts V, van Dijk CN,Kerkhoffs GM (2013) No effect of hole geometry in microfracture for talarosteochondral defects. Clin Orthop Relat Res 471(11):3653–3662

Li Y, Frank EH, Wang Y, Chubinskaya S, Huang HH, Grodzinsky AJ (2013)Moderate dynamic compression inhibits pro-catabolic response of cartilageto mechanical injury, tumor necrosis factor-α and interleukin-6, but accentuatesdegradation above a strain threshold. Osteoarthritis Cartilage 21(12):1933–1941

Luo ZJ, Seedhom BB (2007) Light and low-frequency pulsatile hydrostaticpressure enhances extracellular matrix formation by bone marrowmesenchymal cells: an in-vitro study with special reference to cartilage repair.Proc Inst Mech Eng H J Eng Med 221(5):499–507

Madry H, Grün UW, Knutsen G (2011) Cartilage repair and joint preservation:Medical and surgical treatment options. Deutsches Arzteblatt 108(40):669–677

Madry H, van Dijk CN, Mueller-Gerbl M (2010) The basic science of the subchondralbone. Knee surgery, sports traumatology, arthroscopy: official journal of theESSKA 18(4):419–433

Magnussen RA, Dunn WR, Carey JL, Spindler KP (2008) Treatment of focalarticular cartilage defects in the knee: a systematic review. Clin OrthopRelat Res 466(4):952–962

Mauck RL, Nicoll SB, Seyhan SL, Ateshian GA, Hung CT (2003) Synergistic actionof growth factors and dynamic loading for articular cartilage tissueengineering. Tissue Eng 9(4):597–611

Mauck RL, Soltz MA, Wang CC, Wong DD, Chao PH, Valhmu WB, Hung CT,Ateshian GA (2000) Functional tissue engineering of articular cartilagethrough dynamic loading of chondrocyte-seeded agarose gels. J BiomechEng 122(3):252–260

Meachim G, Stockwell RA (1973) The Matrix. Freeman MA (ed) Adult articularcartilage. Pitman medical, London, pp 1–5.

Miller RE, Grodzinsky AJ, Barrett MF, Hung HH, Frank EH, Werpy NM, McIlwraithCW, Frisbie DD (2014) Effects of the combination of microfracture andself-assembling Peptide filling on the repair of a clinically relevant trochleardefect in an equine model. J Bone Joint Surg Am 96(19):1601–1609

Miyanishi K, Trindade MC, Lindsey DP, Beaupre GS, Carter DR, Goodman SB,Schurman DJ, Smith RL (2006a) Dose- and time-dependent effects of cyclichydrostatic pressure on transforming growth factor-beta3-induced chondrogenesisby adult human mesenchymal stem cells in vitro. Tissue Eng 12(8):2253–2262

Miyanishi K, Trindade MC, Lindsey DP, Beaupre GS, Carter DR, Goodman SB,Schurman DJ, Smith RL (2006b) Effects of hydrostatic pressure andtransforming growth factor-beta 3 on adult human mesenchymal stem cellchondrogenesis in vitro. Tissue Eng 12(6):1419–1428

Mizuno S, Ogawa R (2011) Using changes in hydrostatic and osmotic pressure tomanipulate metabolic function in chondrocytes. Am J Physiol Cell Physiol300(6):C1234–1245

Mow VC, Wang CC, Hung CT (1999) The extracellular matrix, interstitial fluid andions as a mechanical signal transducer in articular cartilage. Osteoarthritisand cartilage / OARS, Osteoarthritis Research Society 7(1):41–58

Nebelung S, Gavenis K, Luring C, Zhou B, Mueller-Rath R, Stoffel M, Tingart M,Rath B (2012) Simultaneous anabolic and catabolic responses of humanchondrocytes seeded in collagen hydrogels to long-term continuousdynamic compression. Annals of anatomy = Anatomischer Anzeiger: officialorgan of the Anatomische Gesellschaft 194(4):351–358

Nebelung S, Gavenis K, Rath B, Tingart M, Ladenburger A, Stoffel M, Zhou B,Mueller-Rath R (2011) Continuous cyclic compressive loading modulatesbiological and mechanical properties of collagen hydrogels seeded withhuman chondrocytes. Biorheology 48(5):247–261

Nicodemus GD, Bryant SJ (2010) Mechanical loading regimes affect the anabolicand catabolic activities by chondrocytes encapsulated in PEG hydrogels.Osteoarthritis Cartilage 18(1):126–137

Okuda Y, Konishi R, Miyata S (2013) Effect of cyclic compressive stimuli onmechanical anisotropy of chondrocyte-seeded agarose gel culture. NihonKikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society ofMechanical Engineers, Part C 79(801):1736–1743

Omata S, Sonokawa S, Sawae Y, Murakami T (2012) Effects of both vitamin C andmechanical stimulation on improving the mechanical characteristics ofregenerated cartilage. Biochem Biophys Res Commun 424(4):724–729

Ortved KF, Begum L, Mohammed HO, Nixon AJ (2015) Implantation of rAAV5-IGF-I Transduced Autologous Chondrocytes Improves Cartilage Repair inFull-thickness Defects in the Equine Model. Molecular therapy: the journal ofthe American Society of Gene Therapy 23(2):363–373

Parkkinen JJ, Ikonen J, Lammi MJ, Laakkonen J, Tammi M, Helminen HJ (1993)Effects of cyclic hydrostatic pressure on proteoglycan synthesis in culturedchondrocytes and articular cartilage explants. Arch Biochem Biophys300(1):458–465

Potier E, Noailly J, Ito K (2010) Directing bone marrow-derived stromal cellfunction with mechanics. J Biomech 43(5):807–817

Saris DB, Dhert WJ, Verbout AJ (2003) Joint homeostasis. The discrepancybetween old and fresh defects in cartilage repair. J Bone Joint Surg85(7):1067–1076

Schulz RM, Bader A (2007) Cartilage tissue engineering and bioreactor systemsfor the cultivation and stimulation of chondrocytes. European biophysicsjournal : EBJ 36(4–5):539–568

Shelton JC, Bader DL, Lee DA (2003) Mechanical conditioning influences themetabolic response of cell-seeded constructs. Cells Tissues Organs175(3):140–150

Shepherd DE, Seedhom BB (1999) The ‘instantaneous’ compressive modulus ofhuman articular cartilage in joints of the lower limb. Rheumatology (Oxford)38(2):124–132

Smith RL, Lin J, Trindade MC, Shida J, Kajiyama G, Vu T, Hoffman AR, van derMeulen MC, Goodman SB, Schurman DJ, Carter DR (2000) Time-dependenteffects of intermittent hydrostatic pressure on articular chondrocyte type IIcollagen and aggrecan mRNA expression. J Rehabil Res Dev 37(2):153–161

Smith RL, Rusk SF, Ellison BE, Wessells P, Tsuchiya K, Carter DR, Caler WE, SandellLJ, Schurman DJ (1996) In vitro stimulation of articular chondrocyte mRNAand extracellular matrix synthesis by hydrostatic pressure. J Orthop Res14(1):53–60

Spiller KL, Maher SA, Lowman AM (2011) Hydrogels for the repair of articularcartilage defects. Tissue Eng Part B Rev 17(4):281–299

Suh JK, Baek GH, Aroen A, Malin CM, Niyibizi C, Evans CH, Westerhausen-Larson A(1999) Intermittent sub-ambient interstitial hydrostatic pressure as a potentialmechanical stimulator for chondrocyte metabolism. Osteoarthritis andcartilage / OARS, Osteoarthritis Research Society 7(1):71–80

Torzilli PA, Bhargava M, Chen CT (2011) Mechanical Loading of Articular CartilageReduces IL-1-Induced Enzyme Expression. Cartilage 2(4):364–373

Torzilli PA, Tehrany AM, Grigiene R, Young E (1996) Effects of misoprostol andprostaglandin E2 on proteoglycan biosynthesis and loss in unloaded andloaded articular cartilage explants. Prostaglandins 52(3):157–173

van Dijk CN, Reilingh ML, Zengerink M, van Bergen CJ (2010a) The natural historyof osteochondral lesions in the ankle. Instr Course Lect 59:375–386

van Dijk CN, Reilingh ML, Zengerink M, van Bergen CJ (2010b) Osteochondraldefects in the ankle: why painful? Knee surgery, sports traumatology,arthroscopy : official journal of the ESSKA 18(5):570–580

Wagner DR, Lindsey DP, Li KW, Tummala P, Chandran SE, Smith RL, Longaker MT,Carter DR, Beaupre GS (2008) Hydrostatic pressure enhances chondrogenicdifferentiation of human bone marrow stromal cells in osteochondrogenicmedium. Ann Biomed Eng 36(5):813–820

Waldman SD, Spiteri CG, Grynpas MD, Pilliar RM, Kandel RA (2004) Long-termintermittent compressive stimulation improves the composition andmechanical properties of tissue-engineered cartilage. Tissue Eng10(9–10):1323–1331

Waters RL, Lunsford BR, Perry J, Byrd R (1988) Energy-speed relationship ofwalking: standard tables. J Orthop Res 6(2):215–222

Zengerink M, Szerb I, Hangody L, Dopirak RM, Ferkel RD, van Dijk CN (2006)Current concepts: treatment of osteochondral ankle defects. FootAnkle Clin 11(2):331–359, vi

What quantitative mechanical loading stimulates in vitro ...

Documents

Transcript of What quantitative mechanical loading stimulates in vitro ...